Voter: Paper beats computers

When will government officials realize that sometimes, paper is a better answer than computers alone? And I ask this as someone who really likes what computer technology can do for collecting, storing, organizing and searching data.

But when you've got poorly designed software and improperly trained users, paper can end up the better choice. Even putting aside security issues surrounding touch-screen voting machines, what about the people who are supposed to be overseeing and operating the devices for the voting public? In many cases, local polling places don't have enough technically savvy people to set up and operate e-voting machines, let alone fix them if something goes awry.

Vote-tallying machines themselves aren't the only problem; computerized voter check-in systems are causing issues as well. In DeKalb County, Georgia, for example, voters complained about long lines and lengthy waits, due in part to snail-like computerized voter check-in, the Atlanta Journal Constitution reports:

"The Inman Middle School polling location experienced 90-minute to one-hour delays because poll registration machines kept crashing, officials said. . . .

"A number of citizens endured the roughly hour-long wait, only to be told after finally reaching the front of the line that they were at the wrong precinct. That wasn't exactly welcome news for people hoping to vote before going to work.

"Others still were simply forced to bide their time as polling officials confirmed their personal information.

Several openly expressed their dissatisfaction.

'I used to live in Massachusetts,' said Sara Thompson, 'and up there, you'd be in and out in under 15 minutes because they would just check you off on a paper roll. It seemed like checking out each person on a list computerized took a lot longer.'

I voted in Massachusetts this morning, and can confirm that despite a heavy turnout, paper check-in was quick, lines were short, and I was in and out in less than 15 minutes.

I also marked my ballot on paper, to be read by an optical scanner, and I'm quite happy about that. Yes, I know that scanners can make mistakes, too. But at least if there's a concern about inaccuracies, a recount is possible. How do you do a recount on an electronic machine with no paper trail?